|Predicting gene function based on fusion events|
Edward R. Winstead
July 9, 2001
Two or more independent genes in one genome may be found as one continuous gene in other genomes. Genes that fuse tend to have related functions but different DNA sequences, according to several recent studies. A new study of 30 sequenced microbial genomes found that most of the fusion events involved genes of the same functional family. Therefore, knowing the function of one gene in a fusion event may be useful in characterizing a gene of unknown function, the researchers propose.
Charles DeLisi, of the Department of Biomedical Engineering, Boston University, led the study. The researchers identified fusion links among genes in 30 sequenced microbial genomes and categorized genes with known functions in 15 broad categories such as transcription and energy production. If an uncharacterized gene was linked by fusion to a gene of known function, they found, the latter provided "a strong hint" regarding the function of the former.
In the study, genes of the same functional family were linked in 72 percent of the fusion events. The background probability that two genes are members of the same category is about 9 percent, according to the study published in Proceedings of the National Academy of Sciences.
Because genes being linked tend to be dissimilar in sequence, the researchers say, fusion-link analysis may complement other strategies for characterizing genes such as sequence similarity searches, DNA microarray expression profiles, and phylogenetic analyses.
"Our results suggest that fusions are nonrandom events and therefore presumably confer a selective advantage," DeLisi and colleagues write. "The fusion of genes thus may reflect a strategy for coping with the increasing complexity of higher organisms by effectively compartmentalizing functionally related proteins."
. . .